The invention relates to a method for disturbance-proof recognition of data contained in data signals wherein a time-domain filter is employed for sampling signals allocated to the data signals during a respectively prescribed time duration following signal edges of the data signals so as to recover data contained in the data signals. The invention also relates to an arrangement for the implementation of this method.
Time-domain filters are employed in many types of signal processing, for example in data transmission or in data storage by means of magnetic or optical recording, in order to eliminate disturbances in the data signals. In the signal processing, analog data signals are frequently converted into binary data signals for further processing. In the conversion, the time-domain filters are utilized for the elimination of the noise signals. The time-domain filters are based on the fact that specific types of disturbances that can occur in the signal processing exhibit a limited time duration. By sampling the signals in intervals that are longer than the maximum duration of these disturbances, a reduction or an elimination of these disturbances is achieved.
In a conversion of the analog data signals into binary data signals, the zero-axis crossings of the analog data signals are usually identified and the binary values of the binary data signals correspond to the polarities of the analog data signals. When, as a consequence of disturbances, additional zero-axis crossings appear, for example, due to the sag of the analog data signals between two zero-axis crossings, additional pulse-shaped, binary data signals occur which can lead to errors in the recovery of the data contained in the data signals. These additional zero-axis crossings can appear precisely after the duration for the sampling which is prescribed by the time-domain filter.